Prepare the corpus
Parse documents, scans, tables, and versions so the system can retrieve useful context rather than a pile of unexamined chunks.
Private RAG / retrieval over company material
Your documents can be searchable without training a model on them.
Private RAG retrieves relevant passages from an approved document set and gives the model only the context needed to answer. We build the source, access, citation, update, and test loop around the client's questions rather than selling a vector database as the product.
Approved corpusSource citationsPermission-awareUpdate loop
Pattern demonstration / source corpus varies
What private RAG actually does
The retrieval path is the product.
A useful system knows what to search, what not to expose, when a source is stale, and how to show the evidence behind an answer.
Find the relevant passage
Use the question, permissions, metadata, and retrieval tests to select context for the answer.
Show where it came from
Return source labels and a clear not-found response so the user can review instead of trusting fluent prose.
The parts that decide quality
RAG is not upload and pray.
Keep versions honest
A new policy or contract version should supersede the old material in a known way, not quietly create conflicting answers.
Retrieve within the boundary
Matter, department, project, or role permissions need to apply before context reaches the model.
Test the questions
A fixed set of known questions exposes missed sources, unsupported answers, stale content, and retrieval gaps.
A private RAG pilot
Start with a corpus someone owns.
Choose the question set
Collect the questions staff ask, the answers they expect, and the source that should support each answer.
Prepare the source
Clean, extract, OCR, label, version, and group the document material before indexing it.
Test retrieval and citations
Check permissions, answer quality, not-found behavior, stale versions, and the evidence returned to users.
Define updates
Set the connector, webhook, scheduled job, or upload process that keeps the last known-good knowledge version intact.
The model is only one part of a private RAG system. The source, permissions, update path, and test set decide whether people can trust it.
Pricing / fixed scope
Know the starting numbers before you ask.
The final quote follows the workflow. Infrastructure and model bills stay on your accounts.
One Workflow
Starting from$3,500 / ₦2.5m
2 to 3 weeksOne workflow for a small team, with training and handover.
See the one-workflow packageFirm-wide
Starting from$7,500 / ₦8m
4 to 6 weeksMultiple sources, roles, integrations, and admin handover.
See the deploymentAnnual support
Starting from$3,000 / ₦1.5m
per yearStandard care for one delivered workflow. Optional. Larger deployments and active monitoring are separately scoped.
See supportArchitecture review from $500. Standard annual support is $3,000 / ₦1.5m per year for one delivered workflow. New workflows, integrations, active monitoring, and infrastructure are separately scoped; infrastructure, model, storage, and messaging bills stay on client accounts. Full pricing and what changes the quote
Why trust Pristine3D?
We build and operate production software.
Pristine3D Ltd builds and operates live digital products, and we run private AI workflows internally as part of our own operations. We scope around your real workflow: the documents you own, the questions your team asks, and the access boundary you approve. Based in Lagos, Nigeria, we work remotely with clients worldwide.
We already run live products
Based in Lagos, Nigeria, Pristine3D currently builds and operates smartcards.ng, venu.ng, photoshoot.ng, and ugc.ng in production.
We start with the actual workflow
One input, one output, one test set, and one person who owns the result. We scope a real workflow instead of a transformation programme.
The boundary stays visible
Cloud, model, storage, and messaging accounts stay in your name. The chosen data path, access rules, test record, documentation, and training are part of the agreed scope.
Straight answers
Common questions.
Is private RAG the same as fine-tuning?
No. RAG retrieves current context at answer time. Fine-tuning changes model behavior for a different purpose and is not the default way to make company documents searchable.
Can it handle scanned PDFs?
Yes, if OCR or vision ingestion is included and tested. Scans, tables, and images raise the scope and evaluation work.
Does RAG guarantee correct answers?
No. A strong setup shows sources, refuses unsupported questions, preserves permissions, and is measured against a fixed test set.
Own your knowledge base
The model is not the product. The knowledge base is.
Documents, the retrieval index, access rules, and the workflows built around them are the asset, and they compound. We deploy so the knowledge base stays yours: on your accounts, in the environment you choose, under access rules your team defines. The model behind the answers is a connector, so the knowledge base moves with you, not with a vendor.
Your corpus, your index
The document store, metadata, and retrieval setup live on accounts you own. No vendor holds the corpus.
Models are swappable parts
Change the model provider, move regions, or go local without rebuilding the knowledge base or the workflow.
The knowledge base is the alpha
Every improvement to the corpus improves the answers, and the improvement stays with you, not with a vendor.
Keep exploring
Related setups.
Start with the corpus
What questions should your documents answer?
Tell us the source collection, users, permissions, update frequency, and examples of the questions that keep returning.